DolbyÆ Weighs In On DTSÆ Digital Surroundô

By Perry Sun, Movie Sound Editor

2-Jan-01

At a recent media-exclusive invitation to their headquarters in San Francisco, DolbyÆ Laboratories issued a four-page report of their tests of DTSÆ Digital Surroundô, using the DTS CAE-4 and CAD-4 professional encoder and decoder, respectively, both of which are now available. They also performed comparisons between Dolby Digital and DTS Digital Surround, using both format's respective professional encoder and decoder. The results published in the report were rather curious and interesting, to say the least.Dolby stated, from their measurements of decoded DTS Digital Surround signals, that at the 1509kbps data rate, performance was ""commensurate with a well-designed audio codec."" However, at 754kbps, the company noted that the upper end of the spectral range was 15kHz. They also reported a slight roll-off in the LFE channel, with a 1dB attenuation at 50dB and 3dB at 90Hz. The document further said that Dolby Digital maintained full 20kHz bandwidth at 448kbps, and 18kHz at 384kbps, with LFE response flat (with less than 0.1dB deviation) to 120Hz.Using standard test material to challenge the capabilities of audio codecs, Dolby's report stated that ""audible artifacts were found to varying degrees in the subjective tests with both coders and at all data rates, but 754kbps DTS exhibited the worst.""Other subjective listening tests were performed at mastering studios in London and Los Angeles, with professional recording engineers as the test subjects. The test material consisted of ""high quality"" music and movie soundtracks. Three such tests were performed, and the results are reproduced here from the report:Studio One1. Dolby Digital 4482. DTS 1509/Dolby Digital 448 (tie)3. DTS 754Studio Two1. Dolby Digital 4482. DTS 15093. Dolby Digital 3844. DTS 754Studio Three1. DTS 15092. Dolby Digital 4483. Dolby Digital 3844. DTS 754Various details about these tests, such as the number of participants, number of observations, names of test materials, whether the tests were blind etc. were not specified in the report.Dolby then offered three factors that could possibly affect perceived differences between the formats as described in the media - differences in output levels, differences in mastering and the use of different mixes.The report is intended to have the reader come away with the belief that Dolby Digital equaled, and actually outperformed DTS, and that perceived distinctions between the two formats, as reported in the press and by consumers over the Internet are not solely attributable to the nature of the audio compression.There is certainly no doubt that such a report on a competing sound format has to be taken with a grain of salt, and that third-party evaluation is needed to repeat the tests.